Annals of Biomedical Engineering, Vol. 32, No. 9, September 2004 (© 2004) pp. 1202–1210 Cardiac Optical Mapping under a Translucent Stimulation Electrode JOY LIAU, 1,3 JOHN DUMAS, 1 DEBORAH J ANKS, 2 BRADLEY J. ROTH, 2 and STEPHEN B. KNISLEY 1 1 The Department of Biomedical Engineering of the School of Medicine, The University of North Carolina at Chapel Hill, NC 27599-7575; 2 The Department of Physics, Oakland University, Rochester, MI 48309-4401; and 3 Present address: The School of Medicine, The University of California at San Diego, CA (Received 9 December 2003; accepted 25 May 2004) Abstract—Major effects of stimulation on cardiac transmem- brane potentials (Vm) are thought to occur under the electrode, however these have not been optically mapped due to blockage of light by electrodes. Here we optically mapped under translucent indium tin oxide (ITO) electrodes in hearts stained with trans- membrane voltage sensitive fluorescent dye, di-4-ANEPPS ex- cited at 488 nm. Emissions in wavelength bands 510–570 nm and >590 nm were similarly affected by changes in ITO transmittance due to electrochemical effects of current at the electrode inter- face. Dual-wavelength ratiometric mapping with the two emis- sion bands revealed Vm under the electrode during plateau-phase stimulation (220 mA). Changes in Vm were heterogeneous under the electrode, and were anisotropic with larger values along the fiber axis. These results explain early excitation sites for suffi- ciently strong diastolic stimulation, and agree with theoretical predictions based on summation of anisotropic effects of point stimulation and a linear 3-d cardiac bidomain computer model. The bidomain model and experiments disagree under the edge of the electrode, where modeled Vm is much larger. Thus, changes in Vm under an electrode are anisotropic with greater Vm in the direction parallel to fibers. Nonlinear effects of stimulation in hearts may limit changes in Vm under the electrode edge. Keywords—Heart, Electrical stimulation, Indium tin oxide, Flu- orescent dye, Ratiometry, Bidomain model. INTRODUCTION Electric stimulation or shock is the only effective therapy available to halt ventricular fibrillation. The mechanisms of stimulation in hearts depend on the changes in transmem- brane potential during the stimulation pulse (Vm). The Vm lead to alterations in states of transmembrane voltage- dependent ion channels responsible for the action po- tential, repolarization, reentry, and defibrillation. 2,7,8,11,23 Theory indicates maximal Vm occur in tissue under an electrode, 10,31 and density of applied current is highest un- der the edge of an electrode. 13,21,35 High defibrillation ef- Address correspondence to Stephen B. Knisley, PhD, Department of Biomedical Engineering, The University of North Carolina at Chapel Hill, CB# 7575, 152 MacNider Hall, Chapel Hill, NC 27599-7575. Electronic mail: knisley@bme.unc.edu ficacy when large electrodes are used that contact much of the heart surface suggests effects under electrodes may be important for defibrillation. 6 It has not been practical to measure the effects under electrodes with conventional mapping methods. Arrays of recording electrodes placed under the stimulation electrode to map extracellular potentials would interfere with the stimulation delivery. Optical mapping with transmembrane potential-sensitive fluorescent dye would be impractical un- der metallic stimulation electrodes that block light. Here, we stimulated the heart with translucent indium tin oxide (ITO) disc electrodes that pass light to and from the heart to allow optical examination of subelectrode effects. Map- ping was performed with a 128-spot laser scanner optical mapping system that recorded di-4-ANEPPS fluorescence emission at two wavelength bands. METHODS Electrode Fabrication The ITO was sputtered to a thickness of 200 nm onto one side of 2 inch × 3 inch borosilicate glass plates. This thickness produced a surface resistance of 15 ohms per unit square region of the plate (a standard unit material resistance). 1 After the ITO was cleaned with acetone and dehydrated, the ITO was patterned to form a bipolar pac- ing electrode pair with separate runs and a single 1-cm diameter ITO disc electrode near the center of the plate (Fig. 1). A 0.75-cm-wide ITO run was used to carry cur- rent between the disc electrode and the top plate edge. The bipolar electrodes were located 2–3 mm from the disc. Wax was deposited on all electrode terminal areas and runs to serve as an etch resist. ITO was then etched for 20 min in solution consisting of 50 parts H 2 O, 50 parts HCl concen- trate (37.6%), and 1 part HNO 3 concentrate (69.5%) at 25 degrees C to remove all ITO not covered with wax. The plate was then washed with warm water followed by ace- tone. Leads were attached to ends of runs near plate edges with conductive cement. Attachment areas and runs were insulated with epoxy cement and urethane enamel. A total 0090-6964/04/0900-1202/1 C  2004 Biomedical Engineering Society 1202